Poly (ADP-ribose) polymerase (PARP) is a nuclear enzyme that catalyzes poly (ADP-ribosyl)ation, which is one of the important post-translational modifications in eukaryotic cells. Since its discovery over 40 years, PARP has attracted many scholars' attention due to its importance in repair of DNA damage and maintenance of genomic stability. In particular, encouraging progress has been made in the relationship between PARP and tumorigenesis and the improvement of tumor therapy by regulating PARP.
Recent studies have revealed that the PARP family comprises at least subtypes including PARP-1, PARP-2, PARP-3, PARP-4/VPARP, tankyrases (TANK-1, TANK-2 and TANK-3). In human genome, a total of 16 different genes encode PARP superfamily members, all of which have a highly conservative PARP catalytic domain consisting of 50 amino acid residues. Except the conservative PARP catalytic domain, the PARP superfamily members have different primary structures, intracellular locations and specific substrates, indicating that poly (ADP-ribose) is a post-transcriptional regulation protein with different biological functions.
PARP-1 is a polypeptide consisting of 1014 amino acid residues with a molecular weight of 113 kDa and an isoelectric point of 8.0˜9.8. It is the first PARP family member discovered and is best characterized in the family. The PARP-1 polypeptide comprises 3 functional domains. (1) An N-terminal DNA binding domain (DBD, 46 kDa) that ranges from amino acid residues 1˜372 comprises a nuclear localization sequence (NLS) and two Zn finger motifs. These two Zn finger motifs participate in the DNA gap recognition. The first Zn finger motif recognizes DNA single and double-strand breaks, and its mutation will greatly reduce the activation of PARP; the second Zn finger motif only participates in the recognition of DNA single-strand breaks. (2) An automodification domain (22 kDa) that ranges from amino acid residues 374-525, through which PARP-1 binds to ADP ribose and becomes self-glycosylated. The automodification domain may also lead to dimerization of PARP-1. (3) A C-terminal catalytic domain (54 kDa) that ranges from amino acid residues 524-1014 provides the basis to convert NAD into an ADP ribose. The sequence of the catalytic domain is highly conservative, especially the segment of amino acid residues 859-908, which is 100% conservative in vertebrate. The catalytic domain consists of two parts, wherein amino acid residues 662-784 form six α-helixes of A, B, C, D, E, F, which is a unique structure of PARP. The helix region binds to the active region, namely the NAD binding site through F α-helix, which probably relates to the transduction of activation signals. In addition, studies have found that PARP also comprises a leucine zipper structure, which is presumed to play a role in the homodimerization or heterodimerization of PARP.
PARP-1 participates in the DNA gap recognition. After the DNA gap recognition, activated PARP-1 forms homodimers and catalyzes the cleavage process of NAD+, which converts into nicotinamide and ADP ribose, the later was used as the substrate for poly-(ADP-ribosyl)ation of nuclear receptor proteins. The activation of PARP-1 is positively correlated with the degree of DNA damage. PARP-1 activity may also be regulated through a negative feedback by self-repair during the poly (ADP-ribosyl)ation induced by DNA damage, forming a DNA damage-stimulated PARP-1 active cycle. Studies on PARP-1 inhibition and knock-out mice confirmed that PARP-1 plays a critical role in the maintenance of genomic stability.
PARP is abundant in human cells, especially in the immune cells and germ cells. Poly (ADP-ribosyl)ation occurs in many physiological processes. The multiple roles of PARP comprise chromatin degradation, DNA replication, DNA repair, gene expression, cell division, differentiation, and apoptosis.
Upon DNA damage, the PARP-1 enzyme is activated and binds to the DNA to catalyze the ribosylation of poly adenosine diphosphate, and thus initiates the process of DNA damage control and repair. On the other hand, overreaction of PARP-1 could lead to the depletion of NAD+/ATP and eventually lead to cell necrosis. The apparently contradictory dual role of PARP-1 has attracted the broad interest of biologists and pharmaceutical chemists. As a result, small molecule inhibitors of PARP-1 have been widely investigated for anti-tumor therapy, nerve injury inhibition and inflammation injury. It is of great significance to discover and seek PARP-1 inhibitors.
At the molecular level, the strategies for most cancer therapies, e.g. radiotherapy and chemotherapy, kill the tumor cells by damaging the cancer cells' DNA. Accordingly, targeted therapies involved in the recognition, reaction and repair of DNA damage have been the research hotspot in recent years. PARP-1 plays an important role in DNA repair, cell death, proliferation and differentiation. It is hypothesized that the inhibition of PARP-1 activity may result in the inhibition of PARP-1-mediated repair of DNA damage, enhancing the damage of radiotherapy and chemotherapy to tumor cell DNA. Therefore, PARP-1 is of potential therapeutic value in tumor therapy.
Currently, a variety of PARP-1 inhibitors, for example, BSI-201, AZD-2281, ABT-888, NU1025, GP115427, AG014699, CEP-6800, AG14361 and INO1001, etc. have entered clinical trials. Among these PARP-1 inhibitors, AZD-2281 (olaparib/KU-59436) developed by AstraZeneca is an oral small molecule inhibitor of PARP-1, mainly used to treat ovarian cancer, breast cancer and solid tumors. In April 2009, AstraZeneca reported that AZD-2281 alone showed significant inhibitory activity for gastric cancer with good tolerance in Phase I clinical trial. At present, the research of AZD-2281 in combination with cisplatin, carboplatin, paclitaxel for solid tumors treatment is in Phase II clinical trial stage, and the clinical study mainly targeting hereditary breast and ovarian cancer has entered Phase III clinical trials. In addition, a novel PARP-1 inhibitor AG014699 developed by Pfizer is a promising malignant melanoma drug in combination with alkylating agent TMZ, and this compound has entered Phase II clinical trials.
The PARP-1 inhibitors, used alone or in combination with other chemotherapy or radiotherapy, have made important progress in the therapeutic studies in molecular, cellular and organism level in chemotherapy. Many studies have found that PARP-1 inhibitors can strengthen the effects of various antineoplastic drugs (e.g. alkylating agents, topoisomerase inhibitors). Tentori et al. (Blood 2002, vol. 99: 2241) found that NU1025 can prolong the survival of the mice bearing brain lymphoma. Delaney et al. (Clin Cancer Res 2000, vol. 6: 2860) used 12 human tumor cell lines (representing lung, colon, ovarian and breast tumor, each comprising 3 cell lines) and found that the combination of PARP-1 inhibitors NU1025 or NU10285 with antineoplastic drugs alkylating agent (temozolomide, TMZ) or camptothecin (topotecan, TP) can enhance their inhibition of tumor cell growth. The tumor-inhibition enhancement effect of NU1025 and NU10285 correlates to their strength of PARP-1 inhibition. The PARP-1 inhibitor GPI15427 developed by Guilford Inc. can enhance the effect of TMZ and significantly prolong the survival of the mice bearing glioblastoma multiforme, cerebral lymphoma or intracranial malignant melanoma. Studying on the mice subcutaneously transplanted with colon cancer cells, Calabrese et al. (J Natl Cancer Inst 2004, vol. 96: 56) found that AG14361 enhanced the antineoplastic effect of TMZ. Miknyoczki et al. (Mol Cancer Ther 2003, vol. 2: 371) treated the xenograft models corresponding to clinically obtained tumor (i.e. U251MG human glioblastoma, HT29 human colon adenocarcinoma and Calu-6 non-small cell lung cancer) with PARP-1 inhibitor CEP-6800 combined with TMZ, CPT-11 (camptothecin) or cisplatin, showing that CEP-6800 can increase the time and/or percentage of the G2/M phase arrest of the tumor cells treated with TMZ, CPT-11 or cisplatin, and enhance the therapeutic effects of TMZ, CPT-11 and cisplatin on the tumor-bearing nude mice.
Although the studies on the combination of PARP-1 inhibitors and chemotherapy showed significant enhancement in anti-tumor effects, people still concerns about its safety and feasibility. Miknyoczki et al. has found that the therapeutic dose of CEP-6800 did not increase the damage of the three chemotherapy drugs to the isolated human intestinal epithelial and kidney cells, or increase the TMZ gastrointestinal toxicity and cisplatin nephrotoxicity to the living mice, which indicates that the combination of CEP-6800 and chemotherapy is valuable and feasible. Moreover, PARP-1 inhibitor GPI15427 can achieve effective concentration in target tissue by oral administration due to its long half-life and high bioavailability, and a combination treatment of GPI15427 and TMZ on models of central nervous system tumor could achieve good therapeutic efficacy.
Some reports demonstrated that the activation of PARP plays a role in cell death in a number of disease states, suggesting that PARP inhibitors would have therapeutic efficacy in those conditions. For example, enhanced poly(ADP-ribosyl)ation has been observed following focal cerebral ischemia in the stroke rat, it indicated that the PARP was activated (Tokime et al. J. Cereb. Blood Flow Metab. 1998, vol. 18, 991). A substantial body of published pharmacological and genetic data supports the hypothesis that PARP-1 inhibitors would be neuroprotective following cerebral ischemia, or stroke. Inhibitors of PARP-1 protected against NMDA- or NO-induced neurotoxicity in rat cerebral cortical cultures (Zhang et al., Science 1994, vol. 263, 687; Eliasson et al. Nature Med. 1997, 3, 1089). it was observed that a series of PARP-1 inhibitors have protective effect on the nerve.
Suto et al. (U.S. Pat. No. 5,177,075) found that the potent PARP-1 inhibitor DPQ (3,4-dihydro-5-[4-(1-piperidinyl)butoxy]-1(2H)-isoquinolinone) provided a 54% reduction in infarct volume in a rat model of focal cerebral ischemia (permanent MCAo and 90 min bilateral occlusion of the common carotid artery) following i.p. dosing (10 mg/kg) two hours prior to and two hours after the initiation of ischemia (Takahashi et al. Brain Res. 1997, vol. 829, 46). Intracerebroventricular administration of a less potent PARP-1 inhibitor, 3-aminobenzamide (3-AB), yielded a 47% decrease in infarct volume in mice following a two hour occlusion of the MCA by the suture thread method (Endres et al. J. Cereb. Blood Flow Metab. 1997, vol. 17, 1143). Treatment with 3-AB also enhanced functional recovery 24 hours after ischemia, attenuated the decrease in NAD levels in ischemic tissues, and decreased the synthesis of poly(ADP-ribose) polymers. Similarly, 3-AB (10 mg/kg) significantly reduced infarct volume in a suture occlusion model of focal ischemia in the rat (Lo et al. Stroke 1998, vol. 29, 830). The neuroprotective effect of 3-AB (3-30 mg/kg, i.c.v.) was also observed in a permanent middle cerebral artery occlusion model of ischemia in the rat (Tokime et al. J. Cereb. Blood Flow Metab. 1998, vol. 18, 991).
The availability of PARP knockout mice (Wang, Genes Dev. 1995, vol. 9, 509) has also helped to validate the role of PARP in neurodegeneration. In the mouse suture thread model of ischemia, an 80% reduction in infarct volume was observed in PARP−/− biallelic knockout mice, and a 65% reduction was noted in PARP monoallelic knockout mice (PARP+/− mice). Endres et al. (1997) found a 35% reduction in infarct volume in PARP−/− mice and a 31% reduction in PARP+/− animals. In addition to neuroprotection, PARP−/− mice demonstrated an improvement in neurological score and displayed increased NAD+ levels following ischemia.
Activation of PARP has been implicated in the functional deficits that may result from traumatic brain injury and spinal cord injury. In a controlled cortical impact model of traumatic brain injury, PARP−/− mice displayed significantly improved motor and cognitive function compared to PARP+/− mice (Whalen et al. J. Cereb. Blood Flow Metab. 1999, vol. 19, 835). Peroxynitrite production and PARP activation have also been demonstrated in spinal cord-injured rats (Scott et al. Ann. Neurol. 1999, vol. 45, 120). These results suggest that PARP inhibitors could avoid the loss of function following head or spinal trauma.
The role of PARP as a mediator of cell death following ischemia and reperfusion may not be limited to the nervous system. In this connection, a recent publication reported that a variety of structurally distinct PARP inhibitors, including 3-AB and related compounds, reduce infarct size following cardiac ischemia and reperfusion in the rabbit (Thiemermann et al. Proc. Nat. Acad. Sci. 1997, 94, 679). In the isolated perfused rabbit heart model, inhibition of PARP reduced infarct volume and contractile dysfunction following global ischemia and reperfusion. Skeletal muscle necrosis following ischemia and reperfusion was also attenuated by PARP inhibitors. Similar cardioprotective effects of 3-AB in a rat myocardial ischemia/reperfusion model were reported by Zingarelli and co-workers (Zingarelli et al. Cardiovascular Research 1997, 36, 205). These in vivo results are further supported by data from experiments in cultured rat cardiac myocytes (Gilad et al. J. Mol. Cell Cardiol. 1997, 29, 2585). Inhibitors of PARP (3-AB and nicotinamide) protected the myocytes from the reductions in mitochondrial respiration observed following treatment with oxidants such as hydrogen peroxide, peroxynitrite, or nitric oxide donors. The genetic disruption of PARP in mice was recently demonstrated to provide protection for delayed cellular injury and production of inflammatory mediators following myocardial ischemia and reperfusion (Yang et al. Shock 2000, 13, 60). These data support the hypothesis that administration of a PARP inhibitor could contribute to a positive outcome following myocardial infarction.
The activity of PARP is also implicated in the cellular damage that occurs in a variety of inflammatory diseases. Activation of macrophages by pro-inflammatory stimuli may result in the production of nitric oxide and superoxide anion, which combine to generate peroxynitrite, resulting in formation of DNA single-strand breaks and activation of PARP. The role of PARP as a mediator of inflammatory disease is supported by experiments employing PARP inhibitors in a number of animal models. The PARP inhibitor 5-iodo-6-amino-1,2-benzopyrone reduced the incidence and severity of arthritis in these animals, decreasing the severity of necrosis and hyperplasia of the synovium. In the carrageenan-induced pleurisy model of acute local inflammation, 3-AB inhibited the histological injury, pleural exudate formation and mononuclear cell infiltration characteristic of the inflammatory process (Cuzzocrea et al., Eur. J. Pharmacology 1998, 342, 67).
Furthermore, PARP inhibitors appear to be useful for treating diabetes. In diabetic patients, the blood glucose concentration maintains at a high level for a long term, which disrupts the endothelial cell stability. For example, hyperglycemia will cause the release of the oxidation medium (e.g. NADH/NADPH oxidase) from the mitochondrial electron-transport chain, and enhance the iNOS expression level, leading to the excessive release of vascular endothelial iNOS. The peroxides and superoxides produced by these processes lead to DNA damage, which activates PARP and depletes cellular NAD+, thus triggers a series of pathological processes that lead to the dysfunction of the entire cell and eventually cell death. Heller et al., “Inactivation of the Poly(ADP-Ribose)Polymerase Gene Affects Oxygen Radical and Nitric Oxide Toxicity in Islet Cells,” J. Biol. Chem., 270:19, 11176-80 (May 1995), discusses the tendency of PARP to deplete cellular NAD+ and induce the death of insulin-producing islet cells. Heller et al. used cells from mice with inactivated PARP genes and found that these mutant cells did not show NAD+ depletion after exposure to DNA-damaging radicals. The mutant cells were also found to be more resistant to the toxicity of NO.
Recent comprehensive review of the prior art has been published by Jagtap and Szabo in Nature 2005, vol 4: 421. Various PARP inhibitors have been reported in the following references, Hattori et al., J Med Chem 2004, vol. 47: 4151 and Menear et al., J Med Chem 2008, vol. 51: 6581. “PARP inhibitor” has also become as the key words for a number of patents including: WO 99/11645, WO 00/32579, WO 02/36599, WO 02/36599, WO 03/103666, WO 03/063874, WO 2004/096779, WO 2005/023246, WO 2005/054210, WO 2006/003148, WO 2007/138355, WO 2008/017883, US 2004/0248931, US 2006/0063926 and US 2007/0093489, etc.